math_indexfandomcom-20200215-history
Idea
you need to aggregate the new master list after each application, otherwise how do you extract it !!?? be sure to manually review batch edits before applying also put in symbol to colorize, type large type set grapheme and see what color washes it seems like colors may relate to syllables hyperlink color names alwasyss sacrificing perfection for expedience make color notation first idea, fast. not time to look at nested, but rather the layer described by title don't remove redundency in master list make sure to even have titles and keywords linked in all articles for reference put entries as enter form lists so that you can double click to highlight include links to keywords in title, since title doesn't get t o be linked what links here feature to save a lot of time. look for other needs as features for now, leave out linking of concepts to graphemes as unnecessary. once you get a static system, consider migrating to a fresh wiki and considering this one a practice how can you start optimizing results to prepare for color tests, better yet, keep final, minimum chapter kernal records, filing them in the chapter ,and also a kernel category and start an ongoing master kernel so that you can begin that process concurrently. use the master kernel for colorization, since it is the only thing that will fit those requirements. and then start a beta implementation of color processes simultaneous to wiki population including more info in proof snapshots is actually better be sure to indicate when a proof is in the article vector convention should be included with proof might be easier to do multiple chapters taking notes, and then multiple chapter kernels, instead of switching back and forth. make kernel computation a second resort by instead just using the largest concept to be widdled away at when making the next smallest concept. this can reduce a 3 kernel computation to 1, or a 2 to 0 etc. then you can atleast use the kernel as a starting point on unfinished pages, leaving the tag unfinished until you do the rest manually bottom to top hyperlinking you can even take snapshot of all concepts first too, and inf which ones are going to be suitable to maybe be in kernel, and do first pass only on those. finish off kernel pages by entering kernel into regex and removing kernel to generate rest of list for reference, remove line breaks try to make things as modular as you can. go for redirect before making a new page be careful to note which exactly concepts have been added to the kernel for efficiency you can also temporarily skip articles that won't be in the kernel possibly, construct kernel alongside note taking, so you can take extra care when matching terminology do a single pass by putting \s command before single character terms, just be sure that none of them are at top of list, otherwise space check breaks. , use two list manipulator tabs to do this. guiding principle, minimize number of key strokes/clicks. try to pull notes with as little descriptive interceding text as possible. also, as soon as an element is linked to another... it doesn't need to happen that both are included in future articles, since you could probably find that out through the net. most important thing is to make terms as consistent as possible, so during note taking, make some hand notes to help keep consistence during the creation process. another important area is to combine articles when possible--- use an appropriate chapter sub-level as a guide generally best practice is to use chapter titles verbatim as much as possible small thing... special characters don't need string demarkation, since they aren't going to show up in words also, click number reduced by doing as much as possible on a page at on time, if time allows question to ask self when taking notes-- is it a title or a description\ instead of just using once per chapter, you can use the kernel several times. doing similar concepts in groups, as a more flexible tool only include pages and keywords for graphemes or titles of graphemes. note: this can include pages that seemingly don't have graphemes, but are actually constructed of just titles.